FEMALE GENITAL
MUTILATION/CUTTING
ISBN-13: 978-92-806-3941-4 ISBN-10: 92-806-3941-2
Cover photo: © UNICEF/HQ01-0099/Stevie Mann
Text: © The United Nations Children’s Fund (UNICEF), 2005
Female genital mutilation/cutting (FGM/C) is a traditional practice with severe health conse-quences for girls and women. It occurs mainly in countries along a belt stretching from Senegal in West Africa to Somalia in East Africa and to Yemen in the Middle East, but it is also practised in some parts of South-East Asia. Reports from Europe, North America and Australia indicate that it is practised among immigrant communities as well. As with many ancient practices, FGM/C is carried out by communities as a heritage of the past and is often associated with ethnic identity. Communities may not even question the practice or may have long forgotten the reasons for it. Since the early 1990s, data on FGM/C have been collected through a separate module of the Demographic and Health Surveys (DHS) implemented by Macro International. The FGM/C module has yielded a rich base of data comparable over 19 countries. We wish to thank the DHS project for making these data available. Data have also been collected through the Multiple Cluster Indicator Surveys (MICS) using a module similar to that of DHS. The MICS FGM/C module has been adjusted to the DHS module and will be implemented during the third round of surveys (MICS-3) in 2005–06.
The prime objective of this publication is to improve an understanding of issues relating to FGM/C in the wider framework of gender equal-ity and social change. FGM/C is a manifestation
of structural inequality and violates universally recognized human-rights principles of equality and non-discrimination.
By providing a statistical analysis and background, this publication also complements the November 2005 Innocenti Digest on FGM/C.
The publication of Female Genital Mutilation/ Cutting: A Statistical Exploration is the result of a cooperative effort between UNICEF’s Strategic Information Section (SIS/DPP) and its Child Protection Section (CP/PDF). The study was coordinated by Edilberto Loaiza (SIS/DPP). Rada Noeva (SIS/DPP consultant) researched and pre-pared a fi rst draft of this paper, and Claudia Cappa actively participated in the fi nal stages (SIS/DPP). Diakathe Ngagne (SIS/DPP) provided support for data processing and tabulations. Valuable insight and comments were provided by Maria Gabriella De Vita (CP/PD). Special thanks to Catherine Langevin-Falcon, the editor.
For further information or to download these and other publications, please visit the website at www.childinfo.org.
Extracts from this publication may be freely repro-duced, provided that due acknowledgement is given to the source and to UNICEF. The Strategic Information Section and Child Protection Section (UNICEF, New York) invite comments on the con-tent and suggestions on how it could be improved as an informational tool.
2005
FEMALE GENITAL
MUTILATION/CUTTING
I. INTRODUCTION. . . 1
A violation of rights . . . 1
Legal instruments: International and domestic. . 2
UNICEF and FGM/C . . . 2
II. OBJECTIVE AND DATA SOURCES . . . 3
III. GLOBAL PREVALENCE RATES. . . 4
Generational trends. . . 7
IV. SOCIO-ECONOMIC AND DEMOGRAPHIC DIFFERENTIALS AFFECTING FGM/C PREVALENCE RATES . . . 7 Age . . . 7 Education . . . 9 Place of residence. . . 9 Religion . . . 10 Ethnicity . . . 11 Household wealth . . . 12 V. ANALYSIS BY TYPE OF PRACTITIONER. . . 13
VI. ANALYSIS BY TYPE OF FGM/C . . . 15
VII UNDERLYING CAUSES AND ATTITUDES . . . 17
Beliefs vs. practice: Support of FGM/C. . . 17
Socio-economic and demographic differentials . . . 19
Women’s empowerment and support for FGM/C . . . 21
Multivariate analysis . . . 26
VIII. CONCLUSIONS AND RECOMMENDATIONS . . . 28
REFERENCES . . . 30
ANNEXES . . . 31
Table 1A: Prevalence (%) of FGM/C among women 15–49 years old, by place of residence and age group . . . 32
Table 1B: Prevalence (%) of FGM/C among women 15–49 years old, by education level and household wealth . . . 33
Table 1C: Prevalence (%) of FGM/C among women 15–49 years old, by ethnicity and religion . . . 34
ANNEXES (continued) Table 2A: Percentage of women 15–49 years old with at least one daughter circumcised, by place of residence and age group . . . 35
Table 2B: Percentage of women 15–49 years old with at least one daughter circumcised, by education level and household wealth. . . 36
Table 2C: Percentage distribution of women 15–49 years old with at least one daughter circumcised, by ethnicity and religion . . . 37
Table 3: Percentage distribution of women 15–49 years old who have undergone FGM/C, by type of practitioner. . . 38
Table 4: Percentage distribution of women with at least one daughter who has undergone FGM/C, by type of practitioner . . . 39
Table 5: Percentage distribution of women who have undergone FGM/C, by type. . . 40
Table 6: Percentage distribution of women with at least one daughter who has undergone FGM/C, by type . . . 41
Table 7A: Percentage of women 15–49 years old who believe FGM/C should continue, by place of residence and age group . . . 42
Table 7B: Percentage of women 15–49 years old who believe FGM/C should continue, by education level and religion . . . 43
Table 8: Percentage of women and men who support FGM/C and reasons for their support . . . . 44
Table 9: Attitudes towards the continuation of FGM/C and variables of women’s empowerment . . . 45
Table 10: Odds ratio: Likelihood of the discontinuation of FGM/C, according to background characteristics . . . 46
Table 11: FGM/C questions included in DHS and MICS questionnaires for women and men . . . 48
FEMALE GENITAL MUTILATION/CUTTING
A Statistical Exploration 2005
CONTENTS
I. INTRODUCTION
“Even though cultural practices may appear senseless or destructive from the standpoint of others, they have meaning and fulfi l a function for those who practise them. However, culture is not static; it is in constant fl ux, adapting and reforming. People will change their behaviour when they understand the hazards and indignity of harmful practices and when they realize that it is possible to give up harmful practices without giving up meaningful aspects of their culture.”
— Female Genital Mutilation, A joint WHO/UNICEF/UNFPA statement, 1997
Female genital mutilation/cutting (FGM/C) is “the partial or total removal of the female external genitalia or other injury to the female genital organs for cultural or other non-therapeutic reasons.”1 It is estimated that more than 130 million girls and women alive today have undergone FGM/C, primarily in Africa and, to a lesser extent, in some countries in the Middle East.
The World Health Organization (WHO) groups FGM/C into four types:
1. Excision of the prepuce [the fold of skin sur-rounding the clitoris], with or without excision of part or the entire clitoris.
2. Excision of the clitoris with partial or total excision of the labia minora [the smaller inner folds of the vulva].
3. Excision of part or all of the external genitalia and stitching or narrowing of the vaginal open-ing (infi bulation).
4. Unclassifi ed, which includes pricking, piercing or incising of the clitoris and/or labia; stretch-ing of the clitoris and/or labia; cauterization by burning of the clitoris and surrounding tissue; scraping of tissue surrounding the opening of the vagina (angurya cuts) or cutting of the vagina (gishiri cuts); introduction of corrosive substances or herbs into the vagina to cause bleeding or to tighten or narrow the vagina; and any other procedure that can be included in the defi nition of female genital mutilation noted above.2
The procedure is generally carried out on girls between the ages of 4 and 14; it is also done to infants, women who are about to be married and, sometimes, to women who are pregnant with their fi rst child or who have just given birth. It is often performed by traditional practitioners, includ-ing midwives and barbers, without anaesthesia, using scissors, razor blades or broken glass. FGM/C is always traumatic. Immediate compli-cations include excruciating pain, shock, urine retention, ulceration of the genitals and injury to adja-cent tissue. Other complications include septicaemia (blood poisoning), infertility and obstructed labour. Haemorrhaging and infection have caused death.
A violation of rights
FGM/C is a fundamental violation of human rights. In the absence of any perceived medical neces-sity, it subjects girls and women to health risks and has life-threatening consequences. Among those rights violated are the right to the highest attain-able standard of health3 and to bodily integrity.4 Furthermore, it could be argued that girls (under 18) cannot be said to give informed consent to such a potentially damaging practice as FGM/C.
FGM/C is, further, an extreme example of dis-crimination based on sex. The Convention on the Elimination of All Forms of Discrimination against Women defi nes discrimination as “any distinction, exclusion or restriction made on the basis of sex which has the effect or purpose of impairing or nullifying the recognition, enjoyment or exercise by women, irrespective of their marital status, on a basis of equality of men and women, of human rights and fundamental freedoms in the political, economic, social, cultural, civil or any other fi eld” (article 1). Used as a way to control women’s sexuality, FGM/C is a main manifestation of gender inequality and discrimination “related to the his-torical suppression and subjugation of women,”5 denying girls and women the full enjoyment of their rights and liberties.
As stated in the Convention on the Rights of the Child, all actions concerning children should be undertaken in the best interests of the child (article 3.1). The Convention further asserts that children should have the opportunity to develop physically in a healthy way, receive adequate medi-cal attention and be protected from all forms of
violence, injury or abuse. While ‘the best interests of the child’ may be subject to cultural interpreta-tion, FGM/C is an irreparable, irreversible abuse and therefore violates girls’ right to protection.
Governments have sometimes been reluctant to address FGM/C. Considered to be a sensitive issue, it has been widely viewed as a ‘private’ act that is carried out by individuals and family mem-bers rather than state actors. But the health and psychological consequences of the practice itself, as well as the underlying causes that reinforce it, make it imperative for societies, governments and the entire international community to take action towards ending FGM/C.
Legal instruments: International
and domestic
At the international level, the human rights implica-tions of FGM/C have been broadly recognized over time. In Vienna in 1993, the UN World Conference on Human Rights called for the elimination of all forms of violence against women to be seen as a human rights obligation. “In particular, the World Conference stresses the importance of working towards the elimination of violence against women in public and private life…and the eradication of any confl icts which may arise between the rights of women and the harmful effects of certain traditional or customary practices.”6
There are many international treaties and conven-tions that call for an end to harmful traditional practices, including the Convention on the Rights of the Child, the Convention on the Elimination of All Forms of Discrimination against Women, and the African Charter on the Rights and Welfare of the Child. A specifi c focus on FGM/C is found in UN General Assembly Resolution 56/128 on Traditional or Customary Practices Affecting the Health of Women and Girls, and in the Protocol on the Rights of Women in Africa (Maputo Protocol), adopted by the African Union in 2003.
Many of the countries where FGM/C occurs have passed legislation prohibiting the practice, and some countries with large immigrant popula-tions – Canada, France, Sweden, Switzerland, the United Kingdom and the United States – have also outlawed it. Some countries have legal clauses granting asylum to women who fear being muti-lated if they return to their country of origin. For
example, Section 273.3 of the Canadian Criminal Code protects children who are ordinarily resident in Canada (as citizens or landed migrants) from being removed from the country and subjected to FGM/C. The effects of domestic laws on FGM/C prevalence levels are largely understudied; as an indicator, they need to be more closely monitored. At the UN General Assembly Special Session on Children in 2002, governments forged a commit-ment to end FGM/C by 2010. In February 2003, 30 African countries vowed to end FGM/C and called for the establishment of an International Day of Zero Tolerance. That pledge was rein-forced in June of that year at the Afro-Arab Expert Consultation, whose Cairo Declaration highlighted the provision of existing legal tools for the preven-tion of FGM/C.
UNICEF and FGM/C
UNICEF fi rst outlined its position on FGM/C in 1979 as a follow-up to the WHO Seminar on Traditional Practices Affecting the Health of Women and Children (Khartoum Seminar): “The health hazards and psychological risks, long term as well as immediate, to young girls as a result of the practice of female excision in its varied forms are a serious source of concern to UNICEF.”7 In 1980, at the Mid-Decade Conference for Women, UNICEF announced that its support to anti-FGM activities was “based on the belief that the best way to handle the problem is to trigger awareness through education of the public, members of the medical profession and practitioners of traditional health care with the help of local collectives and their leaders.”8
UNICEF, the United Nations Population Fund (UNFPA) and WHO in 1997 released a joint state-ment to bring about a substantial decline in FGM/C in 10 years and to end the practice within three generations. The statement calls for a multidisci-plinary approach and emphasizes the importance of teamwork at the national, regional and global levels. It further identifi es the need to educate the public and lawmakers on the importance of ending FGM/C, to tackle FGM/C as a violation of human rights, in addition to being a danger to women’s health, and to encourage every country where it is practised to develop a national, culturally specifi c plan to end FGM/C.
In its Medium-Term Strategic Plans (MTSP) for 2002–2005 and 2006–2009, UNICEF sees pro-tecting children from violence, exploitation and abuse (including FGM/C) as an integral compo-nent for the protection of their rights to survival, growth and development, and consequently to the achievement of several of the Millennium Development Goals.
II. OBJECTIVE AND
DATA SOURCES
The objective of this study is present to estimates of prevalence levels of FGM/C across and within countries, as well as the circumstances surround-ing the practice. The study presents a global assessment of FGM/C levels and examines differentials in prevalence according to socio-economic, demographic and other proximate variables, including type of FGM/C, practitioners and attitudes towards ending the practice. It further seeks to highlight patterns that exist within the data, illustrate how much can be learned by disaggregating variables and suggest how these data can be used to strategically inform program-matic efforts.
The analysis is centred on women aged 15–49 and their daughters and is based on household survey data from Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS). It focuses on national prevalence rates, the distribution of FGM/C within countries and the circumstances surrounding the practice. Attitudes towards female genital mutilation and support for the discontinuation of FGM/C are explored to determine opportunities for programmatic interventions.
DHS and MICS are nationally representative house-hold surveys designed to measure the health and nutritional status of women and children in the developing world. Standard questionnaires cover a wide range of demographic and health indicators and can include special modules on such topics as FGM/C. The respondents are women aged 15–49, except in Egypt, northern Sudan and Yemen, where the sample of respondents includes only women who have been married. (Most of the results from these three countries, however, are not different from those observed for all wom-en.) As of 2005, the FGM/C module has been implemented in surveys in 20 countries (17 in
sub-Saharan Africa plus Egypt, northern Sudan and Yemen). Since some countries have implemented the module more than once, comparable data sources are available for a total of 30 surveys (see Table 11, page 48, for a comparison of the ques-tions included in each country).
The DHS programme is implemented by Macro International, Inc., an Opinion Research Corporation Company (ORC Macro), for the United States Agency for International Development (USAID) and has been collecting comparable data since the 1980s. The MICS methodology was designed to collect data needed for monitoring progress, initially towards the goals of the 1990 World Summit for Children and more recently towards the 2002 ‘World Fit for Children’ goals and the 2015 Millennium Development Goals. The end-decade round of MICS (1999–2001) was conducted in 66 developing countries, primarily by national government ministries, with technical and fi nancial support from UNICEF and other UN agencies. To date, MICS has included an FGM/C module in three countries: the Central African Republic, Chad and northern Sudan.
Together, DHS and MICS allow a comprehensive picture to be constructed of the current global prevalence rates among women and daughters. They provide valid data on the occurrence of FGM/C practice at national and regional levels. The survey results can also suggest correlations between prevalence and ethnicity, religion or other background variables; indicate how the practice is distributed; help identify girls at risk; and enable monitoring trends over time.
The surveys focus on two types of prevalence indicators. The fi rst addresses FGM/C prevalence levels among women and represents the propor-tion of women aged 15–49 who have undergone FGM/C. The second type of indicator measures the status of daughters and calculates the pro-portion of women aged 15–49 with at least one daughter who has undergone genital mutilation or cutting. In Côte d’Ivoire, Kenya, Niger and the United Republic of Tanzania, the surveys capture the status of the oldest daughter only.
Mauritania 71% Egypt 97% Sudan (north) 90% Ethiopia 80% Kenya 32% Senegal 28% Guinea 99% Burkina Faso 77% Côte d’Ivoire 45% Ghana 5% Benin 17% Mali 92% Yemen 23% Chad 45% Nigeria 19% United Republic of Tanzania 18% Niger 5% Eritrea 89% Central African Republic 36% Group 1: 80% or more Group 2: 25% – 79% Group 3: 1% – 24% FGM/C not widely practised No data available
III. GLOBAL PREVALENCE
RATES
FGM/C occurs throughout the world. WHO esti-mates that between 100 million and 140 million girls and women alive today have experienced some form of the practice.9 It is further estimated that up to 3 million girls in sub-Saharan Africa, Egypt and Sudan are at risk of genital mutilation annually.10 As seen in Figure 1 (below), global prevalence rates display signifi cant regional and geographic variations. In north-eastern Africa, prevalence varies from 97 per cent in Egypt to 80 per cent in Ethiopia. In western Africa, 99 per cent of women in Guinea, 71 per cent in Mauritania, 17 per cent in Benin and 5 per cent in Niger have undergone FGM/C. Where data are available for south-eastern Africa, the prevalence rates are relatively lower at 32 per cent in Kenya, for example, and 18 per cent in the United Republic of Tanzania.
Note: This map does not refl ect a position by UNICEF on the legal status of any country or territory or the delimitation of any frontiers. Sources: DHS and MICS.
Using available survey data, countries where FGM/C is practised can be broadly separated into three groups according to prevalence rates. Countries within each group show similari-ties in the way that FGM/C is practised and in inter-regional prevalence variations. This section refl ects the particular context within which FGM/C is practised and attempts to suggest program-matic interventions and strategies to end it. Group 1 is made up of countries where nearly all women have undergone genital mutilation or cutting. The prevalence rates for countries in this group are high, at 80 per cent or more. Data within countries display very little or insignifi cant variation by socio-demographic variables, including geographic location or background characteristics. In this context, programmes to end FGM/C will be most effective if they target women from all regional and socio-economic groups throughout the country.
For the countries in Group 2, the FGM/C preva-lence rates are at intermediate levels of 25 per cent to 79 per cent. The general characteristics of this tier are that only certain ethnic groups within the country practise FGM/C, at varying intensities. Group 3 also consists of countries where only some ethnic groups within the country practise FGM/C. The countries in this group, however, have low national prevalence rates, between 1 per cent and 24 per cent. Strategies designed to end FGM/C in countries included in Group 3 should consider these variations in prevalence rates along ethnic and regional lines. Programmatic interven-tions will be most effective if they are informed by the differing attitudes and practices among the diverse ethnic communities.
For both Group 2 and Group 3, the presentation of data by socio-economic variables can signifi cantly enhance understanding of the practice and provide a valuable entry point for programmatic interventions. The following presents case studies for three countries – Egypt, Kenya and Benin – as they rep-resent each of the three groups.
In Egypt, a country included in Group 1, FGM/C is almost universal among women of reproduc-tive age. According to the most recent DHS data (2003), among women aged 15–49 who are or have been married the prevalence rate is 97 per cent. Estimates of FGM/C prevalence rates obtained from the past three DHS (1995, 2000 and 2003) are virtually constant, indicating the possibil-ity of no change over the past decade. This can
also be explained, however, by the fact that girls in Egypt generally undergo FGM/C between the ages of 7 and 11, and it would therefore take at least one generation for any decline to be refl ect-ed in the data.
Figure 2 (left) illustrates the consistently high distri-bution of FGM/C throughout the country. Because almost all girls in Egypt undergo FGM/C, few differ-ences in prevalence rates can be observed at the regional or educational levels. For example, 95 per cent of women living in urban areas have under-gone genital mutilation/cutting, compared to 99 per cent of women living in rural areas.
In Egypt in 1959, a ministerial decree made FGM/C punishable by fi ne or imprisonment. Later decrees allowed certain forms but punished others. In 1996, a Ministry of Health decree was upheld by Egypt’s highest administrative court. It prohibited all medical and non-medical practitio-ners from conducting FGM/C in public or private facilities, except for medical reasons certifi ed by the head of a hospital’s obstetric department. In 1997, the Court of Cessation upheld this ban.11 Kenya provides an example of a country where FGM/C is practised only among certain ethnic groups and prevalence rates are intermediate (Group 2). According to the 2003 DHS, 32 per cent of women 15–49 years have undergone FGM/C; the 1998 DHS reported a similar fi gure, 38 per cent.12 From a subnational perspective, Kenya reveals signifi cant regional variations (see Figure 3, below), with FGM/C rates ranging from 4 per cent in the FIGURE 2: FGM/C prevalence in Egypt*
Less than 90% 90% – 97.5% More than 97.5%
* Prevalence among women aged 15–49. Data for the Frontier Governorates are from 2000.
Note: This map does not refl ect a position by UNICEF on the legal status of any country or territory or the delimitation of any frontiers.
Source: DHS, 2003.
FIGURE 3: FGM/C prevalence in Kenya*
Less than 25% 25% – 50% More than 50%
* Prevalence among women aged 15 – 49.
Note: This map does not refl ect a position by UNICEF on the legal status of any country or territory or the delimitation of any frontiers.
west to 99 per cent in the north-east. These region-al variations refl ect the presence of diverse ethnic communities. FGM/C prevalence countrywide is nearly universal among women of the Somali (97 per cent), Kisii (96 per cent) and Masai (93 per cent) groups, and signifi cantly lower among Kikuyu (34 per cent) and Kamba (27 per cent) women. Signifi cant ethnic and regional variations can also be observed in the FGM/C status of daughters. While 21 per cent of all women in Kenya report having their eldest daughter mutilated or cut, the rate reported for daughters of Somali women reaches 98 per cent. Among Luhya, Luo and Swahili women, however, less than 2 per cent report having their eldest daughter circumcised.13 Kenya in 2001 adopted a Children’s Code stating that “no person shall subject a child to female circumcision.” There are, how ever, no signifi cant laws making the practice illegal. The Penal Code does contain provisions pertaining to ‘Offences against Person and Health’, which could be appli-cable to instances involving FGM/C.14
Benin is a Group 3 country in which FGM/C affects only a small proportion of the population. According to the 2001 DHS data, 17 per cent of women aged 15 – 49 have undergone some form of genital mutilation or cutting.
FIGURE 4: FGM/C prevalence in Benin*
Less than 25% 25% – 50% More than 50% No data available
* Prevalence among women aged 15–49.
Note: This map does not refl ect a position by UNICEF on the legal status of any country or territory or the delimitation of any frontiers.
Source: DHS, 2001.
FIGURE 5: Prevalence of FGM/C in women and daughters
0 10 20 30 40 50 60 70 80 90 100 Percentage 92 90 89 80 45 32 18 17 54 47 73 58 63 48 32 24 21 23 5 19 71 77 97 99 66 6 7 20 3 10
Guinea Egypt Mali Sudan (north)
Eritrea Ethiopia Burkina Faso
Mauritania Côte d’Ivoire
Kenya Yemen Nigeria United Republic of
Tanzania
Benin Niger Women (aged 15–49) who have
undergone FGM/C
Women (aged 15–49) with at least one daughter who has undergone FGM/C
At the subnational level, however, FGM/C rates vary signifi cantly across regional and ethnic lines (see Figure 4, page 6), ranging from lower than 2 per cent in the Atlantique, Mono and Oueme regions, to as high as 58 per cent in Borgou. FGM/C is highly prevalent among the Peulh (88 per cent), Bariba (77 per cent) and Lokpa and Yoa ethnic groups (72 per cent), but almost non-exis-tent among the Fon (0.3 per cent). FGM/C is not practised by women of the Adja ethnic group. In Benin, similar variations in prevalence by region and ethnicity can be observed among daughters. A total of 6 per cent of women in Benin report having at least one of their daughters circumcised. In the regions of Atacora and Borgou, these num-bers are 21 per cent and 20 per cent, respectively. In the rest of the country, prevalence among daughters is less than 2 per cent. FGM/C status of daughters is highest among the Bariba (30 per cent), Lokpa and Yoa (26 per cent) and Peulh (38 per cent). It is not done to daughters in the Adja ethnic group and is less than 5 per cent among the other ethnic groups.
Generational trends
The differences between the percentage of women aged 15 – 49 who have undergone FGM/C and the percentage of women aged 15 – 49 with at least one daughter circumcised indicate a change in the prevalence of FGM/C: a generational trend towards ending the practice (see Figure 5, page 6). This is of particular importance in countries where the prevalence among women is higher than 75 per cent. In Egypt and Guinea, for example, where almost all women aged 15–49 have undergone FGM/C, only about half of the women indicated that their daughters have undergone FGM/C.
IV. SOCIO-ECONOMIC
AND DEMOGRAPHIC
DIFFERENTIALS
AFFECTING FGM/C
PREVALENCE RATES
Age
Looking at FGM/C distribution by age cohorts can also provide an indication of how the practice has changed over time. Table 1A (page 32), shows prevalence among women 15–49 by fi ve-year age groups. It can be observed that, overall, most coun-tries demonstrate lower FGM/C prevalence levels in the younger age groups (15–19 and 20–24). However, in the four countries with the highest prevalence of FGM/C (Egypt, Guinea, Mali and northern Sudan) very little evidence of change can be found using this method.
Figure 6 (page 8) compares FGM/C prevalence among women aged 30 – 49 with those aged 15 –29 using the ratio of these two percentages. (A ratio value above 1 indicates that FGM/C is more prevalent among the older cohorts, ages 30 – 49.) Younger generations have lower preva-lence of FGM/C in 11 countries – Benin, Burkina Faso, the Central African Republic, Côte d’Ivoire, Ethiopia, Ghana, Kenya, Nigeria, Senegal, the United Republic of Tanzania and Yemen – all with ratios greater than 1.1 and indicating a possible trend of decrease in the practice. For countries with a higher prevalence – Egypt, Guinea, Mali and northern Sudan – the ratio is very close to 1, indicating that FGM/C is constant across ages and is, therefore, constant during the recent past. In Niger, the ratio is below 1, indicating higher FGM/C prevalence among younger generations. Figure 7 (page 8) presents a similar analysis, but this time refers to the percentage of women aged 20–49 with at least one daughter who has under-gone FGM/C. The ratios presented support the hypothesis of a recent decline in the proportion of women with at least one daughter circumcised in Benin, Burkina Faso, Côte d’Ivoire, Egypt, Ethiopia, Guinea, Kenya, Niger, Nigeria, northern Sudan and the United Republic of Tanzania. When the fi ndings from Figures 6 and 7 are combined, it is possible to conclude that the prevalence of
FGM/C is decreasing among all countries studied here except Eritrea, Mali and Mauritania.
Differences in FGM/C status of daughters vary signifi cantly across countries by mothers’ age (see Table 2A, page 35). In northern Sudan, for exam-ple, 92 per cent of women aged 45 – 49 who are or have been married report that at least one of their
daughters has undergone genital mutilation or cut-ting, compared with 15 per cent of women aged 15 –19. In Mauritania, on the other hand, these differences are less profound, and 68 per cent of women aged 45 – 49 have at least one circumcised daughter, compared to 66 per cent of women in the 15 –19 age group. The mother’s age needs to be considered in conjunction with other factors, FIGURE 6: Ratio of FGM/C prevalences, by women’s age (30 – 49 /15 – 29)
Guinea Egypt Mali Sudan (nor th) Eritr ea Ethiopia Burkina Faso Mauritania Chad Côte d’Ivoire Central African Republic
Kenya Senega l Yemen Nigeria United Republic of Tanzan ia
Benin Ghana Niger Cameroon Ratio of prevalences (women aged 30 – 49 /women aged 15 –29) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7
Note: Countries are listed from higher to lower levels of FGM/C among women aged 15 – 49. A ratio of 1.0 indicates that the prevalences in the two groups are equal.
FIGURE 7: Ratio of FGM/C prevalences in daughters, by mothers’ age (30 – 49 / 20 –29)
Sudan (north) 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0
Mali Mauritania Eritrea Guinea Ethiopia Egypt Burkina Faso
Côte d’Ivoire
Yemen Kenya Nigeria United Republic of Tanzania Benin Niger Ratio of prevalences (mothers aged 30 – 49 /mothers aged 20 –29)
Note: Countries are listed from higher to lower levels of FGM/C among daughters. Data are not available for Cameroon, the Central African Republic and Ghana. A ratio of 1.0 indicates that the prevalences in the two groups are equal.
such as age of circumcision, to be able to estab-lish its impact on the FGM/C status of daughters.
Education
DHS and MICS data further allow the presentation of FGM/C prevalence among women according to their educational attainment. Establishing a relationship between a woman’s FGM/C status and her educational level can often be misleading, as FGM/C usually takes place before education is completed and often before it commences. However, as Table 1B, page 33, shows, FGM/C prevalence levels are generally lower among women with higher education, indicating that circumcised girls are also likely to grow up with lower levels of education attainment.
Mothers’ level of educational attainment, more-over, appears to be a signifi cant determinant of the FGM/C status of daughters (see Figure 8, above). It is generally observed that women with higher education are less likely to have circum-cised daughters than women with lower or no formal education. Of the 15 countries with avail-able survey data, 8 present a positive relationship, with ratios of 1.5 or greater; 6 show no difference by levels of education; and in 1, Nigeria, the likeli-hood of having at least one daughter circumcised is greater among women with some education. (Also see Table 2B, page 36.)
While for most countries the impact of mothers’ education on the FGM/C status of daughters is signifi cant, it is less evident in Eritrea, Guinea, Mali, Mauritania, northern Sudan and Yemen. These individual cases require special analysis to account for the differences observed. In the case of Nigeria, for example, one explanation could be that this difference is “due to the confounding factor of ethnicity, because FGC is practiced by Yoruba groups, who are also more likely to be edu-cated. The lack of association in Mali … could be related to the overall low level of education in the country – 81 per cent of respondents have never been to school, and only 7 per cent had at least some secondary education.”15 In Guinea, which has the highest national FGM/C prevalence rate (99 per cent), this fi nding can also be explained by the very small proportion of women in the country with secondary-or-above schooling (5 per cent, compared with 76 per cent of Guinean women who have no education).
Place of residence
Place of residence is another variable that can be expected to be associated with the levels of FGM/C prevalence. In addition to the effects of urban development, prevalence variations based on residence “are probably rooted in such factors as the area’s ethnic composition, neighbouring countries, dominant religious affi liation, and level of urbanization.”16
FIGURE 8: Ratio of FGM/C prevalences in daughters, by mothers’ education (none/some)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
Mali Mauritania Eritrea Sudan (north)
Guinea Ethiopia Egypt Burkina Faso
Côte d'Ivoire
Yemen Kenya Nigeria United Republic of
Tanzania
Benin Niger
Ratio of prevalences (mothers with no education/
mothers with some education)
As shown in Figure 9 (above), there are nine coun-tries in which place of residence does not affect the observed levels of FGM/C (ratios around 1); these countries are also the ones with the high-est observed levels of FGM/C. For another nine countries, the relationship is positive (ratios above 1), indicating that rural women have signifi cantly higher levels of FGM/C than their urban counter-parts. In Nigeria and Yemen the relationship is the opposite, with much higher values among urban residents (ratios below 1).
In addition, some researchers argue that what are observed as clear urban-rural differences “may be somewhat understated due to urban-rural migra-tion. Many of the countries studied are becoming increasingly urbanized. The infl ux of girls from the rural areas, where prevalence levels are generally higher, into urban areas, may obscure the urban-rural differences in prevalence.”17
Similar urban-rural differentials in prevalence could be observed among daughters (see Table 2A, page 35). In a few countries (for example, Kenya, Niger and the United Republic of Tanzania), women living in urban areas are less likely to have circumcised daughters. In Nigeria the FGM/C status of daughters is higher in the urban areas.
Religion
While religion can help explain FGM/C distribution in many countries, the relationship is not consis-tent. In six of the countries where data on religion are available – Benin, Côte d’Ivoire, Ethiopia, Ghana, Kenya and Senegal – Muslim population groups are more likely to practise FGM/C than Christian groups (see Figure 10, page 11). In fi ve countries there seems to be no signifi cant dif-ferences, while in Niger, Nigeria and the United Republic of Tanzania the prevalence is greater among Christian groups.
Looking at religion independently, it is not pos-sible to establish a general association with FGM/C status. The most marked differences can be observed in Benin, Côte d’Ivoire, Ghana and Senegal. In Côte d’Ivoire, for example, 79 per cent of Muslim women have undergone FGM/C, com-pared with 16 per cent of Christian women. This trend is reinforced in the analysis of FGM/C status of daughters (see Table 2C, page 37). In four countries, Muslim women are more likely to have circumcised daughters than women of other religious affi liations. In Ethiopia, Kenya, Niger and the United Republic of Tanzania, prevalence of FGM/C is higher among daughters of Christian women than among daughters of Muslim women. This could be attributed, however, to other factors FIGURE 9: Ratio of FGM/C prevalences among women, by place of residence (rural/urban)
0.0 0.5 1.0 1.5 2.0 2.5
Guinea Egypt Mali Sudan (north
)
Eritrea Ethiopia
Burkina Faso Mauritania Chad
Côte d’Ivoire Central African Republic
Kenya Senegal Yeme n
Nigeria
United Republic of T
anzania Benin Ghana Niger Cam
eroon
Ratio of prevalences (women living in rural areas
/
women lving in urban areas)
such as ethnicity and the overall distribution of the various religious groups within these countries.
Ethnicity
Among all socio-economic variables, ethnicity appears to have the most determining infl uence over FGM/C distribution within a country. As stated by Dara Carr in an analysis of DHS data, “This fi nding is not surprising, because many researchers have noted that FGC prevalence var-ies with ethnicity or that FGC serves as an ethnic marker.”18 In discussing the role of ethnicity, Ellen Gruenbaum writes: “Female circumci-sion practices are deeply entwined with ethnic identity wherever they are found. Understanding this should provide an important insight into the tenacity of the practice and people’s resistance to change efforts, and it can help to explain why the practice may even spread in certain situations.”19 Table 1C, page 34, shows FGM/C prevalence by country among women aged 15 – 49 in the two ethnic groups with the highest prevalence and in the two ethnic groups with the lowest prevalence. The table provides an indication of the range within which FGM/C varies within a country. For
example, the range of FGM/C differences can be as great as 1 per cent to 95 per cent in Kenya, or as little as 94 per cent to 100 per cent in Guinea. The data point to a trend that was observed earlier in the discussion on the different groups of FGM/C prevalence. In countries where FGM/C is practised almost universally (Group 1), varia-tions by ethnicity are insignifi cant compared to countries where prevalence levels are signifi cantly lower. In Eritrea and Mali, for example, more than 80 per cent of women have undergone some form of FGM/C. In all three of these countries, differ-ences in practices between ethnic groups are less signifi cant compared to countries where FGM/C is less widespread and practised only by certain ethnic groups.
Similar variations in the prevalence of FGM/C by ethnicity can be observed among daughters (see Figure 11, page 12). Marked differentials in preva-lence between ethnic groups exist in countries where FGM/C is not widely practised. In countries where genital mutilation/cutting of girls is almost universal (Guinea and Mali), ethnicity has less of a determining impact upon the likelihood of girls having undergone FGM/C.
FIGURE 10: Ratio of FGM/C prevalences among women, by religion (Muslim/Christian)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Guinea Mali Eritrea Ethiopia Burkina Faso Côte d’Ivoire Central African Republic
Kenya Senegal Nigeria United Republic of
Tanzania
Benin Ghana Niger
Ratio of prevalences (Muslim women/Christian women)
Because it allows a more accurate picture of variations within a country, it is important to disag-gregate prevalence rates by ethnicity. As ethnic identity is one mechanism that determines social relations, within the same country there will be young girls who grow up in an environment where all women have undergone FGM/C, and others growing up in an environment where no women have undergone FGM/C. Ethnicity is a signifi cant variable that can inform the design of programmat-ic interventions in accordance with the specifi cs of the target population.
Household wealth
The household wealth index breaks down the population into quintiles ranging from the richest to the poorest. It is constructed using household asset data and principal components analysis. The asset information is collected through DHS and MICS questionnaires and includes household own-ership of a number of consumer items, ranging from a television to a car, along with such dwell-ing characteristics as sanitation facilities used and access to safe drinking water. Each asset is assigned a weight, and individuals are ranked according to the total score of the household in which they reside. This measure could be used
to determine whether there are FGM/C differen-tials among the wealthiest and poorest sectors of society.
While household wealth appears to have some correlation to FGM/C status, such a relationship is not always consistent. Overall, as Figure 12 (below) indicates, FGM/C prevalence seems to decrease among women of richer households. There are different ways of thinking about the wealth status of women. They could be born into wealth or married into it. From that perspective, looking at the importance of wealth quintiles in terms of FGM/C status is more relevant in regard to daughters. Only in Benin, Chad, Kenya and Mauritania did women living in the poorest 60 per cent of households represent a substantially higher prevalence of FGM/C. While in the remain-ing countries with available data the relationship is not signifi cant, in Nigeria women living in the richest 40 per cent of households experienced a greater prevalence of FGM/C.
Figure 13 (page 13) indicates that in Burkina Faso and Eritrea women in the poorest 60 per cent of households more frequently report that FIGURE 11: Ratio of FGM/C prevalences
among daughters, by mothers’ ethnic group
0 10 20 30 40 50 60 Kenya Niger Ethiopia Benin Côte d’IvoireBurkin a Faso Nigeri a Mali Guinea Ratio of prevalences (mother belongs to one of two ethnic groups with highest prevalence/one of two ethnic groups with lowest prevalence)
Note: Countries are listed in descending order of ratio. A ratio of 1.0 indi-cates that prevalences in the two groups are equal.
FIGURE 12: Ratio of FGM/C prevalences among women, by household wealth (poorest 60%/richest 40%) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Guinea Egypt Eritrea Ethiopia
Burkina FasoMauritania
Chad Kenya Yemen Nigeria Benin
Ratio of prevalences (woman belongs to poorest 60%
of households/richest 40% of households)
Note: Countries are listed from higher to lower levels of FGM/C among women aged 15–49. A ratio of 1.0 indicates that prevalences in the two groups are equal.
their eldest daughter has undergone FGM/C than do their counterparts in the richest 40 per cent of households (also see Table 2B, page 36). In Ethiopia and Nigeria, however, the opposite case is observed, and the proportion of women with at least one circumcised daughter increases from poorest to richest quintile. The data in Nigeria may refl ect a proportionately higher FGM/C prevalence among Christian women, who tend to belong to the richer quintiles, as well as a concentration in the south-east and south-west regions, which are economically better off.
V. ANALYSIS BY TYPE OF
PRACTITIONER
An analysis of the type of FGM/C practitioner provides important insights into the context and circumstances surrounding the practice. In the majority of countries, FGM/C is performed by traditional practitioners, including midwives and barbers. (See Figures 14 and 15, page 14, and Tables 3 and 4, pages 38 and 39.) But recent trends show that in some countries, medical per-sonnel are increasingly involved in FGM/C. The shift from procedures done by traditional prac-titioners to procedures that take place in hospitals and health clinics, done by medical professionals
who use surgical instruments and anaesthetics, is often referred to as ‘medicalization’. DHS and MICS data indicate this is particularly the case in Egypt, Guinea, Kenya, Nigeria, northern Sudan and Yemen, where the medicalization of the practice has dramatically increased in recent years. In all these countries one third or more of the women with at least one daughter circumcised indicate that trained health personnel conducted the procedure. In Egypt, for example, 94 per cent of daughters are found to have undergone FGM/C conducted by trained health personnel (a doctor, nurse, midwife or traditional birth attendant), while this was the case for 79 per cent of mothers (see Tables 3 and 4, pages 38 and 39).
The shift towards medicalization can be attributed to early advocacy efforts aimed at ending FGM/C that placed a strong emphasis on the health con-sequences of the procedure. These initiatives undeniably played an important role in raising public awareness of female genital mutilation and the attendant health risks. However, their over emphasis on the health implications – at the expense of placing the practice in the context of a larger human rights violation – has led to a misconception that medicalization decreases the negative health consequences of the procedure, and is therefore a more ‘benign’ form of the practice. UNICEF’s position is that medicalization obscures the human rights issues surrounding FGM/C and prevents the development of effective and long-term solutions for ending it.
Other organizations throughout the world also actively decry medicalization. They base their posi-tion on the grounds that FGM/C is an irreversible procedure that exposes girls to unnecessary health risks with no perceived medical necessity. The World Health Organization, for example, “strongly condemns the medicalization of female genital mutilation, that is, the involvement of health profes-sionals in any form of female genital mutilation in any setting, including hospitals or other health establishments.”20 The involvement of medical pro-fessionals in the practice, in fact, undermines the message that FGM/C remains a discriminatory act of violence that denies women and girls their right to the highest attainable standard of health and and physical integrity. Experience has shown that in addition to endangering advocacy efforts, the medicalization of FGM/C has served to legitimize and perpetuate the practice in some countries (e.g., Egypt and Sudan).
FIGURE 13: Ratio of FGM/C prevalences among daughters, by mothers’ household wealth (poorest 60%/richest 40%)
Burkina Faso 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4
Eritrea Ethiopia Nigeria
Ratio of prevalences (mother belongs to poorest 60%
of households/
richest 40% of households)
Note: Countries are listed from higher to lower levels of FGM/C among daugh-ters. A ratio of 1.0 indicates that prevalences in the two groups are equal.
Type of practitioner for women: Figure 14 (above) illustrates the percentage of women aged 15 – 49 who underwent FGM/C conducted by a traditional practitioner. With the exception of Egypt and Nigeria – where more than 50 per cent of the procedures involve medical person-nel – the majority of circumcisions are done by traditional practitioners. In Benin, Côte d’Ivoire and Mali more than 80 per cent of circumcisions have been conducted by traditional practitioners. Urban women are generally more likely to have been mutilated or cut by a medical practitioner than rural women (see Table 3, page 38).
Type of practitioner for daughters: Signifi cant changes can be observed in the type of FGM/C practitioner from mothers to daughters (see Figure 15, at right). The most signifi cant shift is the substantial increase in the level of medicaliza-tion between generamedicaliza-tions. Survey data indicate that increasing numbers of women in Egypt (94 per cent), Yemen (76 per cent), Mauritania (65 per cent), Côte d’Ivoire (48 per cent), Kenya (46
per cent), Nigeria (36 per cent), northern Sudan (32 per cent) and Guinea (31 per cent) are using trained health personnel to administer the proce-dure to their daughters. In most other countries, however, more than 70 per cent of circumcisions performed on daughters continue to be done by traditional practitioners: Benin (98 per cent), Burkina Faso (97 per cent), Mali (94 per cent), Niger (93 per cent), Ethiopia (92 per cent), Eritrea (84 per cent) and the United Republic of Tanzania (78 per cent).
While a clear shift can be observed in the type of FGM/C practitioner, it is not possible to estab-lish its direct effect on the overall prevalence of the practice. In countries where the shift is quite signifi cant (e.g., Egypt, Kenya and Nigeria) it could be argued that the change in methods indicates a shift in attitudes as well, resulting from increased awareness of the negative consequences of the practice, the effect of programmatic interventions or other factors. Further analysis is required to test the validity of this hypothesis.
FIGURE 14: Percentage of women who underwent FGM/C by a traditional practitioner Percentage 0 10 20 30 40 50 60 70 80 90 100 Benin Côte d’Ivoir e Mali Burkina Faso Niger Guinea Eritrea United Re public of T anzania Maurit ania Nigeria Egypt
FIGURE 15: Percentage of daughters who underwent FGM/C by a traditional practitioner 0 10 20 30 40 50 60 70 80 90 100 Benin Burkina Faso
Mali NigerEthiopia Eritrea
United Republic of T
anzania Guinea Sudan (north) Nigeria Côte d’Ivoire Kenya MauritaniaYe men Egypt Percentage
Note: Refers to women aged 15–49. Countries are listed in descending order.
VI. ANALYSIS BY TYPE
OF FGM/C
Few of the countries reporting on type of FGM/C use the standard WHO defi nitions, thus making a comparison across countries diffi cult. In most countries, the typology is adapted to refl ect the nuances of the local practice (see Tables 5 and 6, pages 40 and 41).
Important insights into FGM/C itself, as well as the medical complications that may arise, can be gained by analysing the type of FGM/C prac-tised. Data on the type of FGM/C performed on daughters tend to be most accurate because the information is obtained directly from the mother. It is often diffi cult to construct estimates on the type of circumcision in different places, however, as classifi cations may not correspond to local con-cepts or terminology. In addition, because FGM/C occurs predominantly in early childhood, girls may have no recollection of the exact procedure. Other challenges in collecting data on the type of circumcision result from diffi culties in establishing how thoroughly survey respondents understood the questions referring to which of the three main types of FGM/C they had undergone.
In the majority of countries that have included questions regarding type of FGM/C, excision of the prepuce (Type 1) is found to be the most common. Only in Burkina Faso is excision of the clitoris (Type 2) found to be most frequent. A large percentage of women who have undergone the excision of part or all of the external genitalia and the stitching/narrowing of the vaginal opening, or infi bulation (Type 3), is observed in two countries: Sudan, 74 per cent, and Eritrea, 39 per cent (see Table 5, page 40).
Data from the surveys allow the identifi cation of variables associated with the type of circumcision girls undergo. Some scholars hypothesize that different types of circumcision practised within a country can often be attached to differences in ethnicity.21 Because ethnic groups are partially defi ned by sharing a similar religion, religious back-ground is also found to be closely associated with the type of circumcision.
Further correlation could be found between moth-ers’ education and daughtmoth-ers’ type of FGM/C. Certain authors hypothesize that daughters of
educated women are slightly more likely to have received less severe forms of FGM/C compared to daughters of uneducated women. This could be supported by the fact that “more educated women may have a better understanding of the potential hazards of circumcision and are less likely to select it for their daughters.”22
Place of residence often acts as a determinant of the type of FGM/C. Urban women are gener-ally more likely to have a daughter with a less severe type of circumcision than rural women. In describing the link between daughters’ type of circumcision and mothers’ education, it is possible to argue that urban women generally tend to have higher educational status than rural women. Data are most consistent on infi bulation because most countries report on it by name or under ‘sewn closed’, which can generally and quite safely be concluded to be infi bulation. The prevalence of infi bulation varies from less than 1 per cent in Egypt to as high as 74 per cent in northern Sudan (see Figure 16, below). Eritrea has the second
FIGURE 16: Percentage of women who have undergone infi bulation
0 10 20 30 40 50 60 70 80 Sudan (north) Eritre a Guinea Mauritania United Republic of T
anzania Nigeria Benin Ethiopia Côte d’
Ivoire Burkina Faso
Mali Egypt
Percentage
highest prevalence of infi bulation. The 2002 DHS survey in Eritrea found 39 per cent of women aged 15–49 to have been subjected to infi bulation. For both Eritrea and northern Sudan, this method varies by residence, religion and ethnicity.
In Eritrea the prevalence of infi bulation differs signifi cantly according to education, ethnicity and religion. Among women who have undergone FGM/C, 58 per cent of women who have no formal education have been infi bulated, compared to 8 per cent of women with secondary educa-tion. There are substantial religious differences as well, and 87 per cent of Muslim women have undergone infi bulation compared to 32 per cent of Catholic women and 10 per cent of Protestant women. In some ethnic groups infi bulation is near-ly universal. More than 90 per cent of circumcised women have been infi bulated among the Hedarib (99 per cent), the Afar, Bilen and Nara (each at 98 per cent) and the Tigre (91 per cent) ethnic groups. Infi bulation is signifi cantly lower among women from the Tigrigna (3 per cent), the Amhara (7 per cent) and the Rashaida (10 per cent) ethnic groups. Infi bulation is also more common among rural women (47 per cent) then among urban women (21 per cent).
Infi bulation is observed much less frequently among daughters than among women aged 15–49 (Figure 17, at right), although it is still at high and very signifi cant levels. In Eritrea and Guinea, more than 30 per cent of the women surveyed indicated that their daughters have undergone infi bulation. Examining the type of FGM/C among daughters in Eritrea helps lead to the conclusion that FGM/C is often a continuation of the experiences of older family members. Some authors suggest that “older women tend to perpetuate particular types of FGM/C because they expect that younger female relatives should have the same life experi-ences they had.”23 The prevalence of infi bulation in Eritrea among daughters mirrors that of the mothers, and further analysis illustrates this point. As the 2002 DHS for Eritrea indicates, the mother’s educational level appears to signifi -cantly impact the likelihood of a daughter being infi bulated. Forty-nine per cent of women with no formal education have at least one daughter who was infi bulated, compared to 7 per cent of women with secondary education. Religion is another variable that substantially infl uences a daughter
having been infi bulated. Some 77 per cent of Muslim women have at least one daughter who has been infi bulated, compared to 41 per cent for Catholic women and 11 per cent for Protestant women. Similarly to the mothers, daughters of women of the Hedarib (100 per cent), Afar (97 per cent) and Nara (91 per cent) ethnic groups are almost universally infi bulated. Infi bulation is signifi -cantly less prevalent among daughters from the Tigrigna (2 per cent) and Rashaida (13 per cent) ethnic groups. Daughters of women living in urban areas (21 per cent) are signifi cantly less likely to have undergone infi bulation than daughters of women living in rural areas (47 per cent).
Comparable patterns can be observed in northern Sudan, which has the highest prevalence rates of infi bulation among women aged 15–49 who are or have been married. (In Sudan, infi bulation is referred to as ‘pharaonic circumcision’.) Although infi bulation has been against the law in Sudan since the 1940s,24 the practice is still very wide-spread, and 74 per cent of circumcised women in northern Sudan have been infi bulated. The practice varies signifi cantly, however, by religion and region. Among those women who have undergone FGM/C, 83 per cent of Muslim women have been infi bulated, compared to 27 per cent of Christian women. In substantial regional differ-ences, the practice is almost universal in most of the north but is less prevalent in Darfur and in the east.
FIGURE 17: Percentage of daughters who have undergone infi bulation
0 5 10 15 20 25 30 35 40 Eritrea Guinea Mauritania Mali Benin Burkina Faso
Nigeria Ethiopia Egypt
Percentage
VII. UNDERLYING CAUSES
AND ATTITUDES
Social scientists have attempted to address the question of why FGM/C continues. In The Female Circumcision Controversy, Ellen Gruenbaum states: “There is no simple answer to this question. People have different and multiple reasons. Female circumcision is practiced by people of many ethnici-ties and various religious backgrounds, including Muslims, Christians, and Jews, as well as follow-ers of traditional African religions. For some it is a rite of passage. For others it is not. Some consider it aesthetically pleasing. For others, it is mostly related to morality or sexuality.”25
The following pages present the existing evidence from household surveys of support for the prac-tice, the reasons given by survey respondents for supporting the practice, and the differences among background variables, including variables associated with the empowerment of women. At the end of this section, results of a multivariate analysis are presented to identify the net effects of key background variables on the probability of FGM/C being supported at the country level. As a social behaviour, FGM/C derives from a complex set of belief systems. In the majority of countries, the practice is supported among both women and men. The motivation for continu-ing the practice is often linked to the perception of specifi c benefi ts. The reasons for practising FGM/C, however, vary signifi cantly within and between countries (see Table 8, page 44). Some of the ‘benefi ts’ attributed to female genital mutila-tion are summarized below.
Custom and tradition/good tradition: When asked what they believed to be the main reason justifying the continuation of FGM/C, the major-ity of women cite ‘custom and tradition’ or that it is a ‘good tradition’ as a reason for their support. In Côte d’Ivoire, Eritrea and Sudan, for example, around 70 per cent of women fi nd custom and tradition to be the most compelling reason justi-fying the continuation of the practice. In Kenya 42 per cent of women and in Nigeria 35 per cent of women believe FGM/C is a good tradi-tion. P. Stanley Yoder suggests that “among the women who think FGM/C should continue, half to two-thirds regard FGM/C as part of their common-sense understanding of what parents should do
for their daughters – that they are doing what they think is appropriate.”26
In addition to the anthropological justifi cation, that women favour FGM/C predominantly because it is viewed to be a ‘good custom or tradition’ can be found in the methodology of the surveys. DHS provides several answers for the respondents to choose from, and perhaps “this response was the simplest and most succinct way that women could summarize all of the positive qualities they associate with cutting.”27 Regardless of the expla-nations provided, it could be concluded that the largest proportion of women who believe FGM/C should be continued support the practice because of custom or tradition.
Religion: A large proportion of women indicate they believe religion requires FGM/C. This is particularly true in countries with high prevalence rates – 70 per cent of women in Mali, 57 per cent in Mauritania, 33 per cent in Yemen and 31 per cent of women in Egypt believe FGM/C is required by religion.
Other reasons: A widespread belief among women who support FGM/C is that the practice preserves a girl’s virginity, protects her from becoming promiscuous and prevents her from engaging in immoral behaviour. In Mauritania 52 per cent of women and in Kenya and Mali 30 per cent of women believe FGM/C should be contin-ued because it ensures a girl’s virginity.
Another reason women use to justify their sup-port for FGM/C is the belief that a girl cannot be married unless she is circumcised. The belief that FGM/C is necessary to ensure better marriage prospects for a daughter is most widespread among women in Côte d’Ivoire (36 per cent), Niger (29 per cent) and Eritrea (25 per cent). Other frequently mentioned reasons include ‘hygiene and cleanliness’ (which refers to aesthetic judge-ments of physical appearance, rather than to a concept of actually being dirty) and beliefs that FGM/C brings greater pleasure to husbands.
Beliefs vs. practice:
Support of FGM/C
Support for the continuation of the practice is not universal, and it tends to vary within and between countries. Figure 18 (page 18) presents the
proportion of women aged 15 – 49 who indicated that the practice should continue. These numbers are shown next to the total prevalence of FGM/C in each country with the intention of comparing the practice of FGM/C with the attitudes towards its continuation in the future.
As Figure 18 illustrates, responses at the country level vary from as low as the 5 per cent of women in Benin who favour the continuation of FGM/C to more than 70 per cent in Egypt, Mali and northern Sudan. It is interesting to note that high levels of support for the continuation of FGM/C are gener-ally observed in countries where the practice is widespread (prevalence greater than 70 per cent). However, the proportion of women 15–49 who support the continuation of FGM/C is systemati-cally and substantially lower than the proportion that has undergone the practice (particularly in Burkina Faso, where these fi gures are 17 per cent versus 77 per cent).
Another method of analysing attitudes is pre-sented by drawing a correlation between levels of support for the practice and the proportion of women with at least one daughter circumcised. Certain social theories claim that behavioural change can be arrived at by inducing change in attitudes. According to these theories, ending FGM/C will not be reached before an attitudinal change takes place among the main decision mak-ers. This section illustrates an interesting paradox: Attitudes do not necessarily bring behavioural change. Gerry Mackie argues that while women may truly oppose FGM/C, they are unable to stop it by themselves, so they continue to practise it. They are caught in a “belief trap” or set of ideas that “cannot be revised because the believed costs of testing the belief are too high.”28 To examine women’s views on FGM/C by com-paring them to practices, Figure 19 (page 19) illustrates the percentage of women with at least one circumcised daughter versus the percent-age of women who believe the practice should FIGURE 18: FGM/C prevalence among women vs. percentage of women
who support the practice
Central African Republic 17 19 36 32 45 80 89 89 92 5 23 71 77 97 99 5 11 22 20 30 59 17 60 79 49 80 71 68 9 21 0 10 20 30 40 50 60 70 80 90 100
Guinea Egypt Mali Sudan Eritrea (north)
Ethiopia Burkina Faso
Mauritania Côte d’Ivoire
Kenya Yemen Nigeria Benin Niger
Percentage
Women (aged 15–49) who have undergone FGM/C
Women (aged 15–49) who indicated that FGM/C should continue
continue. In most countries, high levels of support for the continuation of the practice are closely correlated with high prevalence among daughters. This suggests women who favour its continuation are more likely to have at least one of their daugh-ters circumcised.
An interesting paradox is presented in Burkina Faso, Eritrea and Mauritania. In Burkina Faso (2003), 32 per cent of women reported at least one of their daughters had undergone FGM/C. At the same time, support for the continuation of the practice was found among only 17 per cent of women. One explanation for this is provided by the passage of legislation banning FGM/C in 1996, which included harsh punishment for those involved in the genital mutilation/cutting of women and girls.29
Eritrea provides another interesting case in which levels of support for the practice are signifi cantly lower (49 per cent) than circumcision status levels among daughters (63 per cent). Certain research-ers attribute this sharp decline in support for the practice in Eritrea to the “gains made by women freedom fi ghters during the long-running war with Ethiopia. The participation of women in the war, combined with the stance against cutting taken by the Eritrean People’s Liberation Front, has been linked to greater societal recognition of
gender issues in Eritrea, including reconsideration of such practices as forced marriages and genital cutting.”30
Socio-economic and
demographic differentials
Similarly to national differentials for prevalence rates of FGM/C, levels of support for the practice vary according to socio-economic and demograph-ic factors. This section analyses women’s support for the practice according to the following back-ground characteristics: age, level of education, place of residence, ethnicity and religion. Age: When comparing age cohort values, atti-tudes among women towards the continuation of the practice display generational differences across countries (Figure 20, page 20). In Eritrea in particular, but also in the Central African Republic, Côte d’Ivoire, Ethiopia, Niger and Nigeria, women in the 15 –29 age group are less likely to support the continuation of the practice than women aged 30 – 49 (ratios greater than 1). In the other coun-tries (except Benin and Burkina Faso), support among younger and older women remains con-stant (ratios around 1), especially in countries with the highest prevalence of FGM/C (Guinea, Egypt and Mali). In Benin and Burkina Faso, women in FIGURE 19: FGM/C prevalence among daughters vs. percentage of womenwho support the practice
63 58 48 24 20 80 59 49 79 68 60 71 30 21 54 21 10 32 47 66 73 3 6 5 17 9 20 11 0 10 20 30 40 50 60 70 80 90 100
Mali Mauritania Eritrea Sudan (north)
Guinea Ethiopia Egypt Burkina Faso
Côte d'Ivoire
Yemen
Kenya Nigeria Benin Niger
Percentage
Women (aged 15–49) with at least one daughter who has undergone FGM/C Women aged (15–49) who indicated that FGM/C should continue
